Stephen C. Tweedie <[EMAIL PROTECTED]> wrote:
: That would only complicate things: it would mean we'd have to scan
: both lists on fsync instead of just the one, for example. There are a
we already do; filemap_fdatasync() is called first in sys_fsync(), though
it usually doesn't have much work I guess.
: number of places where we need buffer lists for dirty data anyway,
: such as for bdflush's background sync to disk. We also maintain the
: per-page buffer lists as caches of the virtual-to-physical mapping to
: avoid redundant bmap()ping. So, removing the buffer_heads which alias
: the page cache data isn't an option. Given that, it's as well to keep
: all the inode's dirty buffers in the one place.
keeping dirty pages in the address space list doesn't preclude any of the
above. the pages could still have buffer_heads attached to them, and
these would cache the block location and be a part of the dirty buffer
list used by bdflush.
I guess both approaches would be roughly the same from a performance
point of view. I feel that keeping all data pages in the address space is more
elegant from a design point of view, but that's quite subjective, of course.
ganesh
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/
